The project demonstrates how you can develop applications with Jakarta EE using widely adopted architectural best practices like Domain-Driven Design (DDD). The project is directly based on the well known original Java DDD sample application developed by DDD pioneer Eric Evans' company Domain Language and the Swedish software consulting company Citerus. The cargo example actually comes from Eric Evans' seminal book on DDD. The original application is written in Spring, Hibernate and Jetty whereas the application is built on Jakarta EE.
The application is an end-to-end system for keeping track of shipping cargo. It has several interfaces described in the following sections.
For further details on the project, please visit: https://eclipse-ee4j.github.io/cargotracker/.
The project website has detailed information on how to get started.
The simplest steps are the following (no IDE required):
- Get the project source code.
- Ensure you are running Java SE 8 or Java SE 11.
- Make sure JAVA_HOME is set.
- As long as you have Maven set up properly, navigate to the project source root and
type:
mvn clean package cargo:run - Go to http://localhost:8080/cargo-tracker
To set up in Eclipse, follow these steps:
- Set up Java SE 8 or Java SE 11, Eclipse for Enterprise Java Developers and Payara 5 (any version up to and including 5.194 will work reliably. Later Payara 5 versions have known issues noted below). You will also need to set up Payara Tools in Eclipse.
- Import this code in Eclipse as a Maven project, Eclipse will do the rest for you. Proceed with clean/building the application.
- After the project is built (which will take a while the very first time as Maven downloads dependencies), simply run it via Payara 5.
After the application runs, it will be available at: http://localhost:8080/cargo-tracker/. Under the hood, the application uses a number of Jakarta EE features including Faces, CDI, Enterprise Beans, Persistence, REST, Batch, JSON Processing, Bean Validation and Messaging.
There are several web interfaces, REST interfaces and a file system scanning interface. It's probably best to start exploring the interfaces in the rough order below.
The tracking interface let's you track the status of cargo and is intended for the general public. Try entering a tracking ID like ABC123 (the application is pre-populated with some sample data).
The administrative interface is intended for the shipping company that manages cargo. The landing page of the interface is a dashboard providing an overall view of registered cargo. You can book cargo using the booking interface. One cargo is booked, you can route it. When you initiate a routing request, the system will determine routes that might work for the cargo. Once you select a route, the cargo will be ready to process handling events at the port. You can also change the destination for cargo if needed or track cargo.
The Handling Event Logging interface is intended for port personnel registering what happened to cargo. The interface is primarily intended for mobile devices, but you can use it via a desktop browser. The interface is accessible at this URL: http://localhost:8080/cargo-tracker/event-logger/index.xhtml. For convenience, you could use a mobile emulator instead of an actual mobile device. Generally speaking cargo goes through these events:
- It's received at the origin location.
- It's loaded and unloaded onto voyages on it's itinerary.
- It's claimed at it's destination location.
- It may go through customs at arbitrary points.
While filling out the event registration form, it's best to have the itinerary handy. You can access the itinerary for registered cargo via the admin interface. The cargo handling is done via Messaging for scalability. While using the event logger, note that only the load and unload events require as associated voyage.
You should also explore the file system based bulk event registration interface. It reads files under /tmp/uploads. The files are just CSV files. A sample CSV file is available under src/test/resources/handling_events.csv. The sample is already set up to match the remaining itinerary events for cargo ABC123. Just make sure to update the times in the first column of the sample CSV file to match the itinerary as well.
Sucessfully processed entries are archived under /tmp/archive. Any failed records are archived under /tmp/failed.
Don't worry about making mistakes. The application is intended to be fairly error tolerant. If you do come across issues, you should report them.
All data entered is wiped upon application restart, so you can start from a blank slate easily if needed.
You can also use the soapUI scripts included in the source code to explore the REST interfaces as well as the numerous unit tests covering the code base generally.
As mentioned earlier, the real point of the application is demonstrating how to create well architected, effective Jakarta EE applications. To that end, once you have gotten some familiarity with the application functionality the next thing to do is to dig right into the code.
DDD is a key aspect of the architecture, so it's important to get at least a working understanding of DDD. As the name implies, Domain-Driven Design is an approach to software design and development that focuses on the core domain and domain logic.
For the most part, it's fine if you are new to Jakarta EE. As long as you have a basic understanding of server-side applications, the code should be good enough to get started. For learning Jakarta EE further, we have recommended a few links in the resources section of the project site. Of course, the ideal user of the project is someone who has a basic working understanding both Jakarta EE and DDD. Though it's not our goal to become a kitchen sink example for demonstrating the vast amount of APIs and features in Jakarta EE, we do use a very representative set. You'll find that you'll learn a fair amount by simply digging into the code to see how things are implemented.
Cargo Tracker's testing is done using JUnit and Arquillian. The Arquillian configuration uses a remote container (Payara 5). Therefore, to perform a test you will need to make sure to have a container running.
For testing locally you will first need to run a Payara 5 server.
You can do that with the following script:
wget https://repo1.maven.org/maven2/fish/payara/distributions/payara/5.194/payara-5.194.zip
unzip payara-5.194.zip && cd payara5/bin
./asadmin start-domainNow for running the tests:
mvn clean verify -DskipTests=falseThis project complies with the Google Java Style Guide. You can use the google-java-format tool to help you comply with the Google Java Style Guide. You can use the tool with most major IDEs such as Eclipse and IntelliJ.
In addition, for all XML, XHTML and HTML files we use a column/line width of 100 and we use 4 spaces for indentation. Please adjust the formatting settings of your IDE accordingly.
- If you use a Payara version greater than 5.194, the batch process will not execute reliably. This is due to the fact that newer Payara vesions use H2 instead of Derby and the shift has caused the Jakarta Batch functionality to stop working reliably. This is a known issue that is being looked at.
- You may get a log message stating that Payara SSL certificates have expired. This won't get in the way of functionality, but it will stop log messages from being printed to the IDE console. You can solve this issue by manually removing the expired certificates from the Payara domain, as explained here.
- If you restart the application a few times, you will run into a bug causing a spurious deployment failure. While the problem can be annoying, it's harmless. Just re-run the application (make sure to completely shut down Payara first).
- Sometimes when the server is not shut down correctly or there is a locking/permissions issue, the H2 database that the application uses get's corrupted, resulting in strange database errors. If this occurs, you will need to stop the application and clean the database. You can do this by simply removing ./cargo-tracker-database from the file system and restarting the application. This directory will typically be under $your-payara-installation/glassfish/domains/domain1/config.